B1226
Title: Tight concentration inequality for sub-Weibull random variables with variance constraints
Authors: Heejong Bong - Carnegie Mellon University (United States) [presenting]
Arun Kuchibhotla - Carnegie Mellon University (United States)
Abstract: The advancement of high-dimensional statistical inference has underscored the need for concentration inequalities applicable to a wider array of random variables. The focus is on sub-Weibull random variables, which span heavy-tailed distributions beyond the scope of sub-Gaussian or sub-exponential random variables. A particular case of recent interest involves sub-Weibull random variables with variance constraints. Concentration inequalities are presented, specifically tailored to address this problem. The problem is formulated as an optimization task to obtain a precise upper bound, providing an optimal solution.
